Saturday, August 24, 2013
Wednesday, August 21, 2013
Install Visual Studio Express 2012 for Windows Desktop
Download the TPC-H
Compile the dbgen.exe
Open the tpch.sln from the tpch_2_16_0\tpch_2_15_0\dbgen\ and click the BUILD.
There would be some error, however, the ddlgen.exe still create successfully under Debug folder
Tuesday, August 20, 2013
- Install Centos 6.4
- Download DB2 10.5
- Add the Host Name resolution: /etc/hosts –> This is required because I don’t have DNS setup for my VM.
- libstdc++so.6: I use yum whatprovides libstdc++so.6 and install the missing package –> libstdc++4.4.7-3-el6.i686
- /lib/libpam.so*: This is 32-bit PAM modules. I use yum whatprovides /lib/libpam.so* to search and install the missing package –> pam-1.1.1-l3.el6.i686
- openibd : This is packaged is needed for purescale. The package has been rename to RDMA ( Remote Direct Memory Access) in Centos 6. Since I am not testing purescale, I am going to ignore.
- Disable SELinux : Edit /etc/selinux/config and reboot.
Sunday, August 18, 2013
I had the opportunity to attend the IBM PoT event for DB2 10.5 BLU. One of the big features for the DB2 10.5 release is the BLU Acceleration and how it can improve on the analytical environment .The columnar store data structure is not the new concept among database vendor, however it is interesting to see how IBM implement and position itself among the competitors.
What is BLU stands for ?
The internal Code name for BLU was Blink Ultra. Now it is called BLU.
Currently, it is stands for Big Data,Lightning fast and Ultra-easy.
But you know it can changed any time .
What edition of DB2 10.5 support DB2 BLU Acceleration ?
BLU Acceleration is a feature of DB2 and just like any database features it is only supported on the certain edition. Currently, it is only supported in :
- Advanced workgroup server edition
- Advanced Enterprise Server edition
BLU does not support in the DPF, HADR and purescale at this point. ( 2013 August) , it is in the high priority list within IBM.
Friday, August 9, 2013
RMAN – Recovery Manager
- Backup is happen on the data block level. It can save time and space. Would detect the HWM on the file and do not backup data block higher the HWM.
- Can detect the bad data block
- Backup Level 0-5. 0 is Full backup. 1-5 can be differential or incremental .
-- Create tablepsace to hold repository
CREATE TABLESPACE RMAN
DATAFILE '/u01/app/oracle/oradata/oracledb1/RMAN01.DBF' SIZE 50M REUSE
AUTOEXTEND ON NEXT 64K MAXSIZE 32767M
EXTENT MANAGEMENT LOCAL
SEGMENT SPACE MANAGEMENT AUTO;
-- Create rman schema owner
CREATE USER rman IDENTIFIED BY rman
TEMPORARY TABLESPACE temp
DEFAULT TABLESPACE rman
QUOTA UNLIMITED ON rman;
GRANT connect, resource, recovery_catalog_owner TO rman;
- If SID is not provided, it use the current $ORACLE_SID
- Catalog’s username name and password is the one we create in the above.
- In below screenshot, we only connect to the catalog
In below screenshot, we connect to the $ORACLE_SID database and catalog and then register database.
backup database plus archivelog;
Here is the screenshot of the backup database plus archivelog
backup tablespace <TABLESPACE NAME>
backup tablepsace exmple;
Backup data file
backup datafile [datafile path]
backup datafile '/u01/app/oracle/oradata/oracledb1/example01.dbf';
Backup control file
backup current controlfile;
Differential Incremental backup
- Incremental backup would only backup the changed block since the last level 0 incremental backup.
- If there is no level 0 incremental backup. RMAN will first create level 0 and then create the level 1 backup.
- Oracle would not use the FULL backup as base (parent ) for increment backup ( level 1).
- The level 1 backup would only backup the changes since last level 0 or level 1 backup.
backup incremental level 1 database;
Cumulative Incremental backup
RMAN can also backup Cumulative backup base on the last level 0 backup.
BACKUP INCREMENTAL LEVEL 1 CUMULATIVE database
# Show all backup
# Show/Delete items not needed for recovery